La,25-Dihydroxyvitamin D3 Regulates the Expression of Carbonic

Proc. Natl. Acad. Sci. USA Vol. 87, pp. 6470-6474, August 1990 Cell Biology la,25-Dihydroxyvitamin D3 regulates the expression of carbonic anhydrase II in nonerythroid avian bone marrow cells (myelomonocytes/osteoclasts/macrophages/bone resorption/acid-base regulation) AGNES BILLECOCQ*t, JANET RETTIG EMANUEL*, ROBERT LEVENSON*, AND ROLAND BARON*t§ Departments of *Cell Biology and tOrthopedics, Yale University School of Medicine, P.O. Box 3333, New Haven, CT 06510 Communicated by Robert W. Berliner, May 8, 1990

ABSTRACT la,25-Dihydroxyvitamin D3 [1,25(OH)2D3], such as the kidney tubule intercalated cell (26), the gastric the active metabolite of the steroid hormone vitamin D, is a mucosa oxyntic cell (27), and the osteoclast (21). In humans, potent regulator of macrophage and osteoclast differentiation. defective CAII is associated with cerebral calcification, tu- The mature osteoclast, unlike the circulating monocyte or the bular acidosis, and osteopetrosis (28), providing genetic tissue macrophage, expresses high levels ofcarbonic anhydrase evidence for CAII involvement in bone resorption. An ana- H (CAH). This enzyme generates protons and bicarbonate logue of the human disease has been produced by chemical from water and carbon dioxide and is involved in bone resorp- mutagenesis in the mouse, where CAII deficiency is associ- tion and acid-base regulation. To test whether 1,25(OH)2D3 ated with renal tubular acidosis and vascular calcification but could induce the differentiation of myelomonocytic precursors no major bone defects (29). toward osteoclasts rather than macrophages, we analyzed its We have investigated whether CAII expression is regu- effects on the expression of CAll in bone marrow cultures lated in bone marrow cultures by 1,25(OH)2D3. We find that containing precursors common to both cell types. The expres- the levels of expression of CAII mRNA and protein are sion ofCAU was markedly increased by 1,25(OH)2D3 in a dose- markedly increased by 1,25(OH)2D3 during the in vitro dif- and time-dependent manner. In bone marrow, this increase ferentiation ofnormal bone marrow precursors. This was also occurred at the mRNA and protein levels and was detectable as observed in a transformed myelomonocytic cell line (BM2), early as 24 hr after stimulation. 1,25(OH)2D3 was also found to thereby confirming a recent report in the transformed human induce CAH expression in a transformed myelomonocytic monocytic cell line HL60 (30). These results suggest that avian cell line. These results suggest that 1,25(OH)2D3 regulates 1,25(OH)2D3 acts directly on myelomonocytic precursor cells the level at which myelomonocytic precursors express CAU, an to induce expression of CAII, a protein that is present at high enzyme that is involved in the function ofthe mature osteoclast. levels in the fully differentiated osteoclast. la,25-Dihydroxyvitamin D3 [1,25(OH)2D3] is the active me- MATERIAL AND METHODS tabolite ofthe steroid hormone vitamin D (1). This metabolite Cell Preparations. Bone marrow cells were isolated from regulates the differentiation of osteoclasts and macrophages, the tibias and femurs of White Leghorn chickens (SPAFAS, two cell types that are thought to be derived from a common Norwich, CT) fed a calcium-deficient diet (Purina) for 1-2 myelomonocytic precursor (2-6). 1,25(OH)2D3 induces an weeks after hatching. Cells passed through a 10-,um nylon increase in bone resorption in vivo (1, 7) and in organ cultures filter (Small Parts) to remove mature multinucleated osteo- (8, 9) but has no effect on the resorptive activity of isolated clasts were fractionated on Ficoll-Paque (Pharmacia) and osteoclasts (10, 11). The increase in bone resorption is plated for 16 hr at 5 x 106 cells per ml in alpha minimum believed to be mediated primarily by means of an effect upon essential medium (Sigma) containing heat-inactivated chick- the proliferation and/or differentiation of osteoclast precur- en (2%) and fetal calf (8%) sera (Sigma), 50 units of penicillin sors within the hematopoietic bone marrow (12-15), leading per ml, and 50 ,ug of streptomycin per ml. To deplete these to an increase in the number of mature osteoclasts (7, 9). This preparations of the most mature stromal and myelomono- view is supported by in vitro experiments with normal bone cytic cells (31), nonadherent cells were recovered after a marrow cells or transformed cell lines. First, 1,25(OH)2D3 16-hr culture and replated at 106 cells per ml for up to 6 days. induces the differentiation and maturation of myelomono- The cells were plated in 96-well plates (105 cells per well) for cytic cell lines toward the mononuclear phagocyte rather than ELISA, on glass coverslips in 24-well plates for immunocy- the granulocyte lineage (3, 12, 15-18). Further, 1,25(OH)2D3 tochemistry (5 x 105 cells per well), and in 150-cm2 flasks (2.5 stimulates the formation of multinucleated osteoclast-like x i05 cells per cm2) for immunochemistry or RNA extrac- cells in long-term bone marrow cultures (13, 14, 19, 20) and tion. Cells were grown in the absence or presence of 10-11 to increases the proportion ofthe cells that can resorb bone (14). 10-7 M 1,25(OH)2D3 (provided by Milan Uskokovic, Hoff- Taken together, these results suggest that this steroid hor- man-LaRoche). mone may specifically induce the differentiation of my- ELISA. After 1-6 days of culture, cells were fixed in 3.7% elomonocytic precursors toward the macrophage and/or formaldehyde, treated with 0.01% H202, and preincubated osteoclast lineage. for 1 hr in phosphate-buffered saline (PBS)/3% bovine serum One of the genes that is expressed at high levels in the albumin. Cells were incubated for 2 hr with antibodies (Abs) osteoclast (21-23), but not in peripheral monocytes or mature diluted in PBS/3% bovine serum albumin with saponin macrophages (24), is the gene encoding carbonic anhydrase (0.05%) to permeabilize the cells. After washing, cells were II (CAII). CAII is an enzyme that reversibly generates bicarbonate and protons from carbon dioxide and water (25). Abbreviations: 1,25(OH)2D3, la,25-dihydroxyvitamin D3; CAII, car- This enzyme plays an essential role in acid-secreting cells bonic anhydrase II; Ab, antibody; mAb, monoclonal Ab; LPS, lipopolysaccharide; PMA, phorbol 12-myristate 13-acetate; AMV, avian myeloblastosis virus. The publication costs of this article were defrayed in part by page charge tPresent address: Department of Experimental Pathophysiology, payment. This article must therefore be hereby marked "advertisement" Pasteur Institute, Paris, France. in accordance with 18 U.S.C. §1734 solely to indicate this fact. §To whom reprint requests should be addressed.

6470 Cell Biology: Billecocq et al. Proc. Natl. Acad. Sci. USA 87 (1990) 6471 incubated with peroxidase-labeled goat anti-mouse IgG Ab RPMI 1640 containing 5% heat-inactivated calf serum, 5% (Cappel) diluted 1:1000 in PBS/3% bovine serum albumin, heat-inactivated chicken serum, 50 units of penicillin per ml, washed, and allowed to react with 2 mg of o-phenylenedi- and 50 pug of streptomycin per ml. The cells were cultured for amine per ml (Sigma) and 0.03%1H202. The reaction was 6 days in the absence or presence of 10 Ag of LPS per ml and stopped after 15 min by the addition of H2SO4. Optical 0.25 pug of PMA per ml. 1,25(OH)2D3 (10-8 M) was added to densities (ODs) were read in duplicate on an ELISA reading some cultures at day 0. Cells were processed for immunocy- system (EAR400, Labinstruments) at 492/620 nm. Two tochemistry with the mAb to CANl as described above. monoclonal antibodies (mAbs) were used: mouse anti- chicken CAII as culture supernatant [7C6-1 (32) diluted 1:20] RESULTS and anti-LEP100 as ascites fluid [(33) diluted 1:1000]. The We analyzed the expression of CAII in bone marrow cultures background OD, measured without primary Ab or with an immunoblots with an antiserum that reacts spe- irrelevant Ab [anti-dinitrophenyl (gift of Richard Anderson, by probing University of Texas, Dallas) diluted 1:200], was subtracted cifically with CAII (36). A marked enhancement in the level from the ODs measured with the specific Abs. In three of the immunoreactive 29-kDa CAII band was observed experiments the protein contents were measured simultane- following 6 days of incubation in the presence of 10-9 M ously (34) after lysis in 0.1% SDS. 1,25(OH)2D3 (Fig. 1). In contrast, cells grown in the absence Western Blots. Cells were washed with PBS, pelleted, and of 1,25(OH)2D3 contained either undetectable (Fig. 1A) or lysed at 40C in lysis buffer [150 mM NaCl containing 1% very low (Fig. 1B) levels of CAI. Quantitation of the Triton X-100, 0.2% SDS, and protease inhibitors (1 mM immunoreactive bands indicated that in response to phenylmethylsulfonyl fluoride, 5 mM EDTA, 10 pug of leu- 1,25(OH)2D3 the levels of CAII increased from 4- to >14-fold. peptin per ml, 1 mM dithiothreitol, and 2 mM benzamidine) To determine whether this effect occurred in a dose-depen- in 50mM Tris buffer at pH 7.4]. Proteins were measured (34), dent manner, CAII levels were determined by ELISA, using electrophoresed through SDS/PAGE (10%), transferred to a CAII-specific mAb (32), in cultures treated with various nitrocellulose (35), and incubated with a rabbit anti-chicken amounts of vitamin D3. Expression of CAII was compared to CAII polyclonal Ab (36) or nonimmune serum diluted 1:2000 that of LEP100, a lysosomal membrane protein (33) that was in Tris-buffered saline. Nitrocellulose sheets were incubated found to vary proportionally to protein content and not to for 2 hr with 0.5 ,Ci of "25I-labeled protein A (Amersham; 1 levels of 1,25(OH)2D3 (data not shown). As shown in Fig. 2A, Ci = 37 GBq) and autoradiographed overnight (Kodak X- CAII expression increased in response to increasing amounts Omat-AT). Reactive CAII bands were quantitated with an of 1,25(OH)2D3 (10-11 to 10-7 M), both after 3 and 6 days in automated densitometer (Bioimage, Ann Arbor, MI). culture; 10-11 M 1,25(OH)2D3 was sufficient to induce a Immunocytochemistry. Cells cultured on glass coverslips significant increase in CAII levels after 6 days. A plateau was were fixed for 30 min at 4°C in 3.7% formaldehyde/0.05% reached at 10-8 M 1,25(OH)2D3 with a 3- to 4-fold increase in saponin (added to all solutions), washed, and incubated for 30 CAII levels after 6 days. As shown in Fig. 2B, the time course min in 0.01%1H202. After preincubation for 20 min in 10% of CAII expression in the presence of 10-9 M 1,25(OH)2D3 goat serum, the coverslips were incubated 12 hr with the mAb was also analyzed by ELISA. An increase in CAII protein anti-LEP100 diluted 1:100 or anti-chicken CAII (7C6-1) cul- levels of >2-fold was detectable 24 hr after stimulation. CAII ture supernatant diluted 1:5, followed by the peroxidase- levels increased for up to 6 days, at which point the values labeled rabbit anti-mouse Fab fragments (Biosys, Marne la were -4 times higher than at day 1. The expression of the Coquette, France) diluted 1:100. Cells were counterstained enzyme also increased in control cultures at day 6, probably with 0.1% toluidine blue in 0.1% sodium borate. Cultures due to the presence of low concentrations of 1,25(OH)2D3 in were also stained for fluoride inhibitable nonspecific esterase the medium (-10-11 M). These results show that the increase as described (4). in expression of CAII is induced by 1,25(OH)2D3 in a dose- Northern Blots. Total RNA was extracted from 6-day-old and time-dependent fashion and is detectable soon (=24 hr or cultures by a modification of the Chirgwin method (37), less) after administration of the hormone. Further, increased separated by electrophoresis through a 1% formaldehyde/ expression of CAII was concomitant with a marked dose- agarose gel, and transferred to Zetabind membranes (Cuno) incorporation (data not by blotting. cDNAs were labeled with [32P]dCTP by the dependent decrease in [3H]thymidine random-priming method (38). Blots were probed with a A B 1.2-kilobase cDNA probe encoding the chicken CAII (39). kDa 1 2 3 4 1 2 3 4 205 - - 40 Hybridization was performed by incubating the labeled cD- _ NAs (0.5-1.0 x 106 cpm/ml) at 42°C overnight in Thomas buffer containing 50% formamide, 50 ,ug/ml each of poly(A), 116 - poly(I), and poly(C) nucleotides (Boehringer Mannheim), 97- C". and 8% dextran sulfate. The membranes were washed in 0.1 x SSCPE (1 x SSCPE = 15 mM NaCl/1.5 mM sodium citrate/ 1.3 mM KH2PO4/100 ,uM EDTA) with 0.1% SDS for 1 hr at 65°C and autoradiographed. Autoradiographic films were 29- t b quantitated by a computerized densitometer scanner (Bio- image). For reprobing, the blots were washed two times at 95°C for 15 min in 0.1x SSCPE/0.1% SDS. They were FIG. 1. Immunoblot analysis of CAII expression in bone marrow successively hybridized with cDNA probes encoding chicken mononuclear cell lysates. Total lysates from 6-day cultures of bone LEP100 (40), human a-actin (gift of Larry Kedes, Stanford, marrow mononuclear cells grown in the absence (lanes 1 and 2) or in CA), and chicken genomic aA globin DNA (41). the presence (lanes 3 and 4) of 10-9 M 1,25(OH)2D3 were separated Culture of Avian Myeloblastosis Virus (AMV)-Transformed on 10% SDS/PAGE and allowed to react either with nonimmune Line BM2. BM2 a line of chicken serum (lanes 1 and 3) or with the polyclonal antibody against chicken Myelomonocytic Cell cells, CAII (lanes 2 and 4). The 29- to 30-kDa band corresponding to CAII myeloblasts transformed with the AMV (clone C3A) (42), is undetectable in lanes 1 and 3. In lanes incubated with the immune were differentiated into macrophages by stimulation with serum, the CAII band either is undetectable (A, lane 2) or is present lipopolysaccharide (LPS; endotoxin of Salmonella typhimu- at low levels (B, lane 2) in the absence of vitamin D3. In contrast, the rium) and phorbol 12-myristate 13-acetate (PMA; Sigma) (43). CAII band is markedly enhanced (4- to >14-fold) in the presence of BM2/C3A cells were plated at 1.0 x 106 cells per ml in medium vitamin D3 (A and B, lanes 4). 6472 Cell Biology: Billecocq et al. Proc. Nati. Acad. Sci. USA 87 (1990)

IA Dav6

P. ~200-

A ..~100

0 0 10-11 10-10 10-9 10-8 10-7 1,25 Dihydroxyvitamin D3 t I concentration (log Molar) *0 .. ,A 300 - 0 OB + Vit D3 (nM) B a

200 FIG. 3. Immunocytochemical localization of CAII in bone marrow mononuclear cells cultured for 6 days in the absence (A) or in the presence (B) of 1,25(OH)2D3 (10-9 M). CAII was localized with the mAb followed by peroxidase-conjugated secondary antibodies; nuclei were counterstained with toluidine blue. (A) In the absence of 9z 100 vitamin D3, CAII is mostly undetectable in the cytoplasm of the cultured cells (identified by their stained nuclei); the closed arrsws point to slightly positive mononuclear and binucleated cells but other binucleated cells are negative (open arrow). (B) In the presence of 1,25(OH)2D3, the level of expression of CAII is increased in most 0 cells, with a gradient of intensity; the closed arrows point to strongly 0 1 2 3 4 5 6 7 positive mononuclear and multinucleated cells; the open arrows Days in Culture point to less positive multinucleated cells. (x360.)

FIG. 2. ELISA analysis of CA11 expression in bone marrow myelomonocytic cell line could also be induced to express * mononuclear cells cultured for 1-6 days in the presence ofincreasing CAII in the presence of vitamin D3. The AMV-transformed concentrations of 1,25(OH)2D3. ODs, measured in 4-12 experiments myeloblastic cell line BM2/C3A (42) was cultured in the (each with duplicate wells), are expressed as a ratio of CA11 to presence of LPS and PMA to induce cell attachment and LEP100 expression (means SEM). (A) The level of expression of differentiation into more mature bone marrow macrophages CA11, measured after 3 days (c) or 6 days (n) in culture, increased (43). Under these conditions, the cells exhibited only mod- as a function of 1,25(OH)2D3 concentration. The first significant erate levels of reaction with CAII antibodies (Fig. 4A). effects are seen at 10-1o M 1,25(OHhD3 after 6 days and 10-9 M after 3 days. A plateau is reached at 10-i M 1,25(OH)2D3 at both time However, when the cells were cultured for 6 days in the points. (B) At all time points, the level ofexpression ofCA11 is higher in cells cultured in the presence of 10-9 M 1,25(OH)2D3 (x) than in cells cultured without 1,25(OH)2D3 (E). CA11 expression is already significantly increased (-3-fold) after 24 hr ofculture with vitamin D3 relative to unstimulated cultures. The expression of CA11 also '9. increases in unstimulated cultures by day 6, probably due to the presence of 1,25(OH)2D3 in the culture medium (about 10-11 M).

shown), consistent with the view that 1,25(OH)2D3 affects the differentiation of cultured bone marrow cells. To determine whether all cells present in the bone marrow A cultures respond uniformly with respect to CAII induction or whether a specific subset of cells is responsive, we analyzed CAII expression in individual cells by immunocytochemis- .1 I try. As shown in Fig. 3A, CAII expression was virtually undetectable in the absence of vitamin D3. In contrast, many of the cells cultured in the presence of 10-9 M 1,25(OH)2D3 were reactive with the CAII mAb (Fig. 3B). Approximately 70%o of the cells exhibited detectable immunoreactivity, but with a gradient of staining intensity. The cells that were most intensely reactive appeared to represent either mononuclear B cells or multinucleated cells containing low (2-10) numbers of nuclei. Most of the very large multinucleated cells (>25 FIG. 4. Immunocytochemical localization of CAII in the AMV- cell line BM2 cultured in the nuclei) were not immunoreactive or exhibited very low levels transformed avian myelomonocytic presence of LPS and PMA and in the absence (A) or presence (B) of of reactivity with the CAII mAb. A majority of the cells was 1,25(OH)2D3 (10-8 M). (A) In the absence of vitamin D3, CAII levels also positive for nonspecific ebterase, suggesting that the were low but still detectable in the cells induced by PMA. (B) In the cultures were highly enriched for mononuclear phagocytes. presence of 1,25(OH)2D3, the level of CAII expression was markedly We therefore investigated whether a chicken bone marrow increased. (X800.) Cell Biology: Billecocq et al. Proc. Natl. Acad. Sci. USA 87 (1990) 6473 presence of 10-8 M 1,25(OH)2D3, an intense staining was the kidney intercalated cell (26), and the osteoclast (21, 22), observed (Fig. 4B). More immature cells, grown in the cell types that are all involved in acid secretion (26, 27, 44). presence of vitamin D3, but in the absence of LPS and PMA Our results indicate that 1,25(OH)2D3 induces the expression induction, did not exhibit an increase in CAII expression. of CAII in myelomonocytic cells. Although the osteoclast is These results suggest that the induction of CAII expression a member of this cell lineage (2, 4-6), other myelomonocytic in bone marrow cultures by 1,25(OH)2D3 occurs in my- cells, including monocytes and macrophages, express only elomonocytic precursors and only when cells progress low levels of CAII (24). The specific induction of CAII through their differentiation program. 1,25(OH)2D3 therefore expression in bone marrow nonerythroid cells suggests that appears to be a potent inducer of myelomonocytic cell 1,25(OH)2D3 plays an important role in the differentiation of differentiation, possibly committing these cells toward the myelomonocytic precursors and their commitment toward osteoclast lineage. the osteoclast lineage. A number of studies have shown that Northern blot analysis was used to analyze CAII mRNA 1,25(OH)2D3 stimulates the formation of osteoclast-like cells expression in vitamin D3-treated cells. Mononuclear cells in long-term bone marrow cultures (13, 19, 20) and can induce isolated from bone marrow were exposed to concentrations the complete differentiation of some of these cells into of 10-11 to 10-8 M 1,25(OH)2D3 for 6 days. As shown in Fig. functional osteoclasts (14). Further, we have shown that late 5, 1,25(OH)2D3 stimulated CAII mRNA levels in a dose- mononuclear precursors ofthe osteoclast express high levels dependent fashion. In contrast, unstimulated cultures con- of CAII (45) and that expression of CAII in short-term bone tained only low levels of CAII mRNA. To ensure that equal marrow cultures is concomitant with that of other markers of amounts of RNA were loaded, the blots were sequentially the mature osteoclast (46), such as the Na',K+-ATPase (47) probed with cDNA encoding LEP100 and a-actin. Actin and the osteoclast functional antigen, a vitronectin receptor mRNA decreased in response to 1,25(OH)2D3, whereas (48). Interestingly, we detect significant changes in the LEP100 mRNA levels remained constant (Fig. 5). By nor- expression of these markers of osteoclast differentiation malizing to LEP100 mRNA, we calculated that CAII mRNA within 24 hr of exposure to 1,25(OH)2D3. In contrast, phe- levels increased from 1.6 times at 10-11 M 1,25(OH)2D3 to 7.2 notypic and functional changes characteristic of osteoclast times at 10-8 M. These results show that treatment of bone differentiation have been reported to require weeks in culture marrow mononuclear cells with 1,25(OH)2D3 results in a (13, 14). Taken together, our results suggest that 1,25(OH)2D3 specific increase in CAII mRNA levels. The regulation of regulates the expression ofcarbonic anhydrase as well as that CAII expression by 1,25(OH)2D3 therefore occurs at pre- of other osteoclast markers (46), inducing myelomonocytic translational levels. cells in the bone marrow to differentiate toward osteoclasts. Northern blots were also probed with cDNA encoding As reported for newborn monocytes (20), we did not, how- aA-globin. No hybridization could be detected in the bone ever, observe actual bone resorption (in the form of resorp- marrow cultures when control erythroid cells were positive tion lacunae) when bone marrow mononuclear cells were (data not shown). This result suggests that 1,25(0H)2D3 cultured on devitalized bone slices, even after 2 weeks in the induction of CAII is specific for the nonerythroid cell pop- presence of 1,25(OH)2D3 (unpublished observations). In con- ulation and does not occur in cells of the erythroid lineage. To trast, formation of bone resorbing osteoclasts was obtained determine whether the levels of CAII expression in these with cultures of total bone marrow (14), including stromal cultures could be affected by other members of the steroid cells, which are an important source of colony-stimulating family of hormones, we examined the effects of retinoic acid, factors (49). 1,25(OH)2D3 alone is therefore most likely not a steroid-like factor known to interact with steroid hormone- sufficient to induce the formation of functional cells, even in responsive elements. Retinoic acid (10-7 to 10-9 M) had no the presence of bone matrix. detectable stimulating effects on the levels of CAII mRNA or Important issues in osteoclast differentiation are whether protein expression (data not shown). This result suggests that these cells share a common precursor with monocytes and 1,25(0H)2D3 is a specific inducer of CAII expression in macrophages and at what stage(s) of differentiation mono- nonerythroid bone marrow cells. nuclear phagocytes can be committed to the osteoclast phe- DISCUSSION notype. Although the cultures we have used are purified by sedimentation and adherence, they are relatively heteroge- CAII is an enzyme that is involved in acid-base regulation. neous for cell types. It was therefore not possible to directly It is expressed at high levels in the gastric oxyntic cell (27), determine if a particular cell type specifically expressed CAII in response to 1,25(OH)2D3. Many cells reacted positively 1,25(OH)2D3 with monoclonal antibodies to CAII after stimulation with 0 lo0"o0 lo' 10i8 1,25(OH)2D3, albeit with a gradient of intensity. A number of lines of evidence, however, suggest that the responsive cells CAll 0P UJJ~I are likely to be committed to the mononuclear phagocytic lineage. First, the isolation procedures and culture conditions LEP100 4_ U fiviI' that we have used are known to markedly enrich for cells in this lineage (5, 13, 31, 35). Second, 1,25(OH)2D3 also induced the expression of CAII in a homogeneous marrow-derived aY-Ac~tin Mt lhE avian myelomonocytic cell line (BM2), a result consistent with the recent report of Shapiro et al. (30) in the transformed FIG. 5. Northern blot analysis ofCAII mRNA expression in bone human promonocytic cell line HL60. Therefore, both trans- marrow mononuclear cells. Bone marrow mononuclear cells were formed and nontransformed bone marrow myelomonocytic cultured for 6 days in the absence or in the presence of increasing cells respond to vitamin D3 with an increase in CAII expres- concentrations of 1,25(OH)2D3. Total mRNA was extracted, elec- sion. trophoresed on agarose gels, blotted to Zetabind membranes, and Regarding the second issue, a number of independent hybridized with radiolabeled cDNA probes encoding the chicken at a inter- CAII; after stripping, the membranes were reprobed with radiola- observations would suggest that it is only stage beled cDNA probes encoding LEP100 and a-actin. A marked dose- mediate between the most immature myelomonocytic pre- dependent increase in CAIl mRNA expression was observed in the cursors and the more mature circulating monocytes that cells presence of 1,25(OH)2D3 [from 1.6 times LEP100 values at 10-11 M can respond to 1,25(OH)2D3 and, possibly, further differen- to 7.2 times at 10-8 M 1,25(OH)2D3]. In contrast, LEP100 mRNA tiate into osteoclasts. First, we report here that BM2 cells can levels remained stable and actin mRNA levels decreased. express CAII at high levels only in the presence of both LPS 6474 Cell Biology: Billecocq et al. Proc. Natl. Acad. Sci. USA 87 (1990) and PMA and 1,25(OH)2D3. In the absence of 1,25(OH)2D3, 10. Chambers, T. J., McSheehy, P. M., Thomson, B. M. & Fuller, K. (1985) BM2 cells differentiate into macrophages (43) but do not Endocrinology 116, 224-239. 11. De Vernejoul, M. C., Horowitz, M., Demignon, J., Neff, L. & Baron, R. express high levels of CAII. Reciprocally, treatment of cells (1988) J. Bone Min. Res. 3, 69-80. with 1,25(OH)2D3 alone (10-8 M) did not induce the expres- 12. Koeffler, H. P., Armatruda, T., Ikekawa, N., Kobayashi, Y. & DeLuca, sion of detectable levels of CAI. This would suggest that H. F. (1984) Cancer Res. 44, 5624-5628. these myelomonocytic cells must pass through an LPS/PMA 13. Roodman, G. D., Ibbotson, K. J., MacDonald, B. R., Kuehl, T. J. & sensitive step in order to acquire CAII responsiveness to Mundy, G. R. (1985) Proc. Natl. Acad. Sci. USA 82, 8213-8217. 14. Takahashi, N., Yamapa, H., Yoshiki, S., Roodman, G. D., Mundy, vitamin D3. Second, mature macrophages and peripheral G. R., Jones, S. J., Boyde, A. & Suda, T. (1988) Endocrinology 122, blood monocytes do not express this enzyme, whereas more 1373-1382. immature monocytes isolated from newborn umbilical chord 15. McCarthy, D. M., San Miguel, J. F., Freake, H. C., Green, P. M., Zola, do (24). Third, only early stages of differentiation in the H., Catovsky, D. & Goldman, J. M. (1983) Leukemia Res. 7, 51-55. monocytic series form multinucleated cells (19, 50) and 16. Abe, E., Miyaurm, C., Sakagamni, H., Takeda, M., Konno, K., Yamazaki, T., Yoshiki, S. & Suda, T. (1981) Proc. Natl. Acad. Sci. USA 78, express the osteoclast vitronectin receptor in response to 4990-4994. 1,25(OH)2D3 (20). 17. Amento, E. P., Bhalla, A. K., Kurnick, J. T., Kradin, R. L., Clemens, The induction of CAII mRNA expression by 1,25(OH)2D3 T. L., Holick, S. A., Holick, M. F. & Krane, S. M. (1984)j. Clin. Invest. suggests that this steroid hormone may regulate transcription 73, 731-739. of the gene. We therefore 18. Mangelsdorf, D. J., Koeffler, H. P., Donaldson, C. A., Pike, J. W. & CAII analyzed the promoter region Haussler, M. R. (1984) J. Cell Biol. 98, 391-398. of the CAII gene (39) for putative, consensus sequences for 19. Takahashi, N., Mundy, G. R., Kuehl, T. J. & Roodman, G. D. (1987) J. vitamin D-responsive elements (VDRE). This analysis did Bone Min. Res. 2, 311-317. not indicate the presence of a region identical to the VDRE 20. Orcel, P., Bielakoff, J. & De Vernejoul, M. C. (1989)Anat. Rec. 225,1-9. described for the rat or the human osteocalcin 21. Gay, C. V. & Mueller, W. J. (1974) Science 183, 432-434. genes (51, 52). 22. Vaananen, H. K. & Parvinen, E.-K. (1983) Histochemistry 78, 481-485. However, there are imperfect palindromic sequences in the 23. Vaananen, H. K. (1984) Histochemistry 81, 485-487. promoter region of the chicken CAII gene [GGTGA (G) 24. Sundquist, K. T., Leppilampi, M., Jarvelin, K., Kumpulainen, T. & TGAAC and GGTGA (C) TCAAC], which are very similar to Vaananen, H. K. (1987) Bone 8, 33-38. reported 25. Maren, T. H. (1%7) Physiol. Rev. 47, 595-781. the VDRE and are compatible with steroid hor- 26. Brown, D., Roth, J., Kumpulainen, T. & Orci, L. (1982) Histochemistry mone-responsive elements (53) and located in positions com- 75, 209-213. patible with the activation of gene transcription (700-750 27. Machen, T. E. & Paradiso, A. M. (1987) Annu. Rev. Physiol. 49, 19-33. base pairs upstream from the TATA boxes). It will clearly be 28. Sly, W. S., Hewett-Emmett, D., Whyte, M. P., Yu, Y. S. & Tashian, of to R. E. (1983) Proc. Nat{. Acad. Sci. USA 80, 2752-2756. interest determine whether the complex formed by 29. Spicer, S. S., Lewis, S. E., Tashian, R. E. & Schulte, B. A. (1989) Am. 1,25(OH)2D3 with its receptor binds to and directly activates 1. Pathol. 134, 947-954. the chicken CAII gene. 30. Shapiro, L. H., Venta, P. J., Yu, Y. L. & Tashian, R. E. (1989) FEBS In conclusion, our results suggest that vitamin D3 acts Lett. 249, 307-310. 31. Beug, H., von Kirchbach, A., Doderlein, G., Conscience, J. F. & Graf, directly on myelomonocytic cells to induce the expression of T. (1979) Cell 18, 375-390. carbonic anhydrase. This gene is expressed at high levels in 32. Linser, P. J., Perkins, M. S., Fitch, F. W. & Moscona, A. A. (1984) the fully differentiated osteoclast and its late precursors and Biochem. Biophys. Res. Commun. 125, 690-697. is necessary for its function in bone resorption (28). 33. Lippincott-Schwartz, J. & Fambrough, D. M. (1986) J. Cell Biol. 102, 1593-1605. 34. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J. (1951) J. We are very grateful to the following scientists for their invaluable Biol. Chem. 193, 265-275. help during this research project: Dr. J. Dodgson for the chicken 35. Towbin, H., Staehelin, T. & Gordon, J. (1984) Mol. Cell. Biol. 4, genomic CAII DNA, the chicken a-globin genomic DNA, the 2587-2593. sequence of the promoter region of the chicken CAII gene, and 36. Linser, P. & Moscona, A. A. (1981) Proc. Natl. Acad. Sci. USA 78, helpful discussions; Dr. Paul Linser for the polyclonal and mono- 7190-7194. clonal antibodies to chicken CAII; Drs. C. and G. Moscovici for the 37. Chomczynski, P. & Sacchi, N. (1987) Anal. Biochem. 162, 156-159. retrovirus-transformed cell 38. Feinberg, A. & Vogelstein, B. (1984) Anal. Biochem. 137, 266-269. myelomonocytic line BM2; and Dr. D. 39. Yoshihara, C. M., Federspiel, M. & Dodgson, J. B. (1984) Ann. N.Y. Fambrough for the monoclonal antibody and the cDNA for LEP100. Acad. Sci. 429, 332-334. We are also very grateful to Sirpa Jamsa-Kellokumpu and Lynn Neff 40. Fambrough, D. M., Takeyasu, K., Lippincott-Schwartz, J. & Siegel, for their technical help. This work was supported by grants from the N. R. (1988) J. Cell Biol. 106, 61-67. National Institutes of Health (DE-04724) and the National Os- 41. Dodgson, J. & Engel, J. D. (1983) J. Biol. Chem. 258, 4623-4629. teoporosis Foundation (AS-101) to R.B. and a grant from the 42. Gazzolo, L.,Moscovici, C., Moscovici, M. G. & Samarut, J. (1979) Cell National Institutes of Health (HL-39263) to R.L., who is an Estab- 16, 627-638. lished Investigator of the American Heart Association. A.B. was 43. Symonds, G., Klempnauer, K. H., Evan, G. I. & Bishop, J. M. (1984) Mol. Cell. Biol. 4, 2587-2593. supported by fellowships from the Pasteur Institute, the Fondation 44. Baron, R., Neff, L., Louvard, D. & Courtoy, P. J. (1985)J. Cell Biol. 101, pour la Recherche Medicale, and the North Atlantic Treaty Orga- 2210-2222. nization. 45. Prallet, B., Male, P., Neff, L., Billecocq, A. & Baron, R. (1990) Proceedings of the 10th International Conference on Calcium Reg. 1. Brommage, R. & DeLuca, H. F. (1985) Endocrinol. Rev. 6, 491-511. Horm. (Elsevier, New York), in press. 2. Ash, P., Loutit, J. F. & Townsend, K. M. S. (1980) Nature (London) 46. Billecocq, A., Emanuel, J., Jamsa-Kellokumpu, S., Prallet, B. Levenson, 283, 669-670. R. & Baron, R. (1990) Proceedings of the 10th International Conference 3. Bar-Shavit, Z., Teitelbaum, S. L., Reitsma, P., Hall, A., Pegg, L. E., on Calcium Reg. Horm. (Elsevier, New York), in press. Trial, J. & Kahn, A. J. (1983) Proc. Nati. Acad. Sci. USA 80, 5907-5911. 47. Baron, R., Neff, L., Roy, C., Boisvert, A. & Caplan, M. (1986) Cell 46, 4. Baron, R., Neff, L., Tran Van, P., Nefussi, J.-R. & Vignery, A. (1986) 311-320. Am. J. Pathol. 122, 363-378. 48. Davies, J., Warwick, J., Totty, N., Philp, R., Helfrich, M. & Horton, M. 5. Burger, E. H., van der Meer, J. W. M., Van de Gevel, J. S., Bribnau, (1989) J. Cell Biol. 109, 1817-1826. J. C., Thesingh, C. W. & van Furth, R. (1982) J. Exp. Med. 156, 49. Spooncer, E., Heyworth, C. M., Dunn, A. & Dexter, T. M. (1986) 1604-1614. Differentiation 31, 111-118. 6. Scheven, B. A. A., Visser, J. W. M. & Nijweide, P. J. (1986) Nature 50. Takahashi, N., Kukita, T., MacDonald, B. R., Bird, A., Mundy, G. R., (London) 321, 79-81. McManus, L. M., Miller, M., Boyde, A., Jones, S. J. & Roodman, G. D. 7. Tinkler, S. M. B., Williams, D. M., Linder, J. E. & Johnson, N. W. (1989) J. Clin. Invest. 83, 543-550. (1981) J. Anat. 133, 91-97. 51. Morrison, N. A., Shine, J., Fragonas, J. C., Verkest, V., McNemeny, 8. Raisz, L. G., Trummel, C. L., Holick, M. F. & DeLuca, H. F. (1972) M. L. & Eisman, J. A. (1989) Science 246, 1158-1161. Science 175, 768-769. 52. Kerner, S. A., Scott, R. A. & Pike, J. W. (1989) Proc. Natl. Acad. Sci. 9. Hermann-Erlee, M. P. M. & Gaillard, P. J. (1978) Calcif. Tissue Res. 25, USA 86, 4455-4460. 111-118. 53. Beato, M. (1989) Cell 56, 335-344.